柔軟なテーブル抽出が必要: Form パーサーは、テーブルのような単純な（行や列にまたがるセルがない）テーブルから抽出します。トレーニングは不要です（また、トレーニングを行うこともできません）。トレーニング済みのテーブル抽出の場合、カスタム抽出ツールは、列（セル）の子フィールドを含む親フィールドで使用できます。
このクイックスタートでは、Document AI の Form Parser 機能について説明します。このクイックスタートでは、 Google Cloud コンソールを使用して Google Cloud プロジェクトと承認を設定し、Form Parser を作成して、Document AI に PDF フォームの処理をリクエストします。
学習内容:
Google Cloud プロジェクトで Document AI を有効にします。
Form パーサー プロセッサを作成します。このプロセッサで、さまざまな種類のドキュメント内のテキスト、Key-Value ペア、テーブル、汎用エンティティを識別し、抽出することができます。
このプロセッサを使用して、サンプル ドキュメントにアノテーションを付けます。
このタスクを Google Cloud コンソールで直接行う際の順を追ったガイダンスについては、[ガイドを表示] をクリックしてください。
Sign in to your Google Cloud account. If you're new to
Google Cloud,
create an account to evaluate how our products perform in
real-world scenarios. New customers also get $300 in free credits to
run, test, and deploy workloads.
In the Google Cloud console, on the project selector page,
select or create a Google Cloud project.
[[["わかりやすい","easyToUnderstand","thumb-up"],["問題の解決に役立った","solvedMyProblem","thumb-up"],["その他","otherUp","thumb-up"]],[["わかりにくい","hardToUnderstand","thumb-down"],["情報またはサンプルコードが不正確","incorrectInformationOrSampleCode","thumb-down"],["必要な情報 / サンプルがない","missingTheInformationSamplesINeed","thumb-down"],["翻訳に関する問題","translationIssue","thumb-down"],["その他","otherDown","thumb-down"]],["最終更新日 2025-09-04 UTC。"],[[["\u003cp\u003eForm Parser is a pre-trained tool that extracts key-value pairs, tables, selection marks, generic entities, and text from documents, suitable for structured forms and flexible table extraction.\u003c/p\u003e\n"],["\u003cp\u003eIt can parse and return data for 11 different generic entities, such as email, phone, address, person, organization, price and date_time, and can also extract data from checkboxes, and tables.\u003c/p\u003e\n"],["\u003cp\u003eForm Parser supports over 200 languages and offers feature support in eight regions, but it cannot be up-trained, and it cannot process radio buttons, nor extract data from blank or empty values.\u003c/p\u003e\n"],["\u003cp\u003eYou can create a Form Parser processor through the Google Cloud console, and then test it using the test document option with a sample PDF file to see the resulting extractions.\u003c/p\u003e\n"],["\u003cp\u003eThe Form Parser tool is designed to be efficient, eliminating the need to build and maintain custom extraction parsers for various high volume form document processing.\u003c/p\u003e\n"]]],[],null,["# Form Parser\n\nProcess documents with Form Parser\n==================================\n\nForm Parser extracts key-value pairs (KVPs), tables, selection marks (like checkboxes),\ngeneric fields, and text to augment and automate document processing.\n| **Note:** Form Parser is pre-trained and cannot be up-trained.\n\nForm Parser can be considered over the other parsers when the use case involves:\n\n- Dealing with structured forms: It excels at extracting KVPs from well-defined forms that look like conventional forms with labeled blanks to fill in, such as `name: __`. Form Parser's pre-trained model offers high accuracy for common fields like names, dates, and addresses.\n- Flexible table extraction is needed: Form Parser extracts from simple (no cells that span rows or columns) tables that look like tables. No training is needed (nor possible). For trained table extraction, the custom extractor can be used with a parent field containing column (cell) child fields.\n- Need efficiency: Avoid building and maintaining extraction parsers, especially for high-volume and varied forms of extraction tasks.\n\nData-extraction features\n------------------------\n\nForm Parser features encompass:\n\n- **KVP:** These are sets of two items within a document---a label or key and its\n corresponding data (a value). You can directly use KVPs (if the keys are consistent)\n or build custom logic to resolve varied keys into consistent structured information.\n\n- **Generic entities:** Parse 11 different fields from documents out of the box. These include:\n\n - `email`\n - `phone`\n - `url`\n - `date_time`\n - `address`\n - `person`\n - `organization`\n - `quantity`\n - `price`\n - `id`\n - `page_number`\n- **Text and layout:** Use our latest OCR engine to extract text and layout\n information. This includes embedded text from digital PDFs (v2.1 only) or text from images.\n\n- **Tables:** Detect and extract tables from images and PDFs.\n\n- **Checkboxes:** A high-quality selection mark detector, which extracts checkboxes\n from images and PDF output as KVP, using the text nearest the checkbox, with a `valueType`\n indicating whether it is filled or unfilled.\n\nLanguages and regions\n---------------------\n\n- Form Parser 2.0 supports over 200 languages. [Learn more](/document-ai/docs/processors-list#expandable-1).\n- We provide feature support in eight regions. [Learn more](/document-ai/docs/regions).\n\nModel versions\n--------------\n\nThe following processor versions are compatible with this feature. For more\ninformation, see [Managing processor versions](/document-ai/docs/manage-processor-versions).\n\nLimitations\n-----------\n\n- Prior JPEG compressions for TIFF are unsupported. Type of JPEG encapsulation defined by the TIFF [version 6.0 specification](https://gitlab.com/libtiff/libtiff/-/commit/f0a54a4fa0cfa377f493d57ee2af393005d5bbe5).\n\n- The checkbox model doesn't support parsing radio buttons. Some detected checkboxes might not have corresponding keys.\n\n- The model doesn't reliably parse a KVP with an unfilled value, such as a blank form.\n\n- The KVP parsing on documents in certain languages may have lower quality than Latin languages.\n\n\u003cbr /\u003e\n\nProcess documents with Form Parser\n----------------------------------\n\nThis quickstart introduces you to the Form Parser feature in Document AI. In this quickstart,\nyou use the Google Cloud console to set up your Google Cloud project and\nauthorization, create a Form Parser, and then make a request for\nDocument AI to process a PDF form.\n\nLearn how to:\n\n1. Enable Document AI in a Google Cloud project.\n\n2. Create a Form Parser processor, which can identify\n and extract text, key-value pairs, tables, and generic entities from many types of documents.\n\n3. Use the processor to annotate a sample document.\n\n*** ** * ** ***\n\nTo follow step-by-step guidance for this task directly in the\nGoogle Cloud console, click **Guide me**:\n\n[Guide me](https://console.cloud.google.com/ai/document-ai?tutorial=document-ai--documentai_form_parser_console)\n\n*** ** * ** ***\n\n- Sign in to your Google Cloud account. If you're new to Google Cloud, [create an account](https://console.cloud.google.com/freetrial) to evaluate how our products perform in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n | **Note**: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Document AI API.\n\n\n [Enable the API](https://console.cloud.google.com/flows/enableapi?apiid=documentai.googleapis.com)\n\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n | **Note**: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Document AI API.\n\n\n [Enable the API](https://console.cloud.google.com/flows/enableapi?apiid=documentai.googleapis.com)\n\nCreate a Form Parser processor\n------------------------------\n\nUse the Google Cloud console to create a Form Parser processor. See [creating and managing processors](/document-ai/docs/create-processor) for more information.\n\n1. In the Google Cloud console navigation menu, click **Document AI** and\n select **Processor Gallery**.\n\n [Processor\n Gallery](https://console.cloud.google.com/ai/document-ai/processor-library)\n2. In the **Processor Gallery** ,\n search for\n **Form Parser** and select **Create**.\n\n\n3. In the side window, enter a **Processor name** , such as `quickstart-form-processor`.\n\n4. Select the region closest to you.\n\n5. Click the **Create** button.\n\nYou're taken to the **Processor Details** page of your new form parser processor.\n\nTest processor\n--------------\n\nAfter creating your processor, you can send annotation requests to it.\n\n1. [Download the sample document](https://storage.googleapis.com/cloud-samples-data/documentai/GeneralProcessors/FormParser/intake-form.pdf).\n\n It's a PDF file containing a sample handwritten medical intake form. This document is stored in a publicly accessible Cloud Storage bucket.\n2. Click the\n **Upload Test Document** button and select the document you just downloaded.\n\n3. You should now be on the **Form Parser analysis** page. You can view the OCR detected text, key-value pairs, tables, and generic entities extracted from the document.\n\n\nClean up\n--------\n\nTo avoid unnecessary Google Cloud charges, use the\n[Google Cloud console](https://console.cloud.google.com/) to delete your processor and [project](https://console.cloud.google.com/cloud-resource-manager) if you don't need\nthem.\n\nWhat's next\n-----------\n\n- Review the [Processors list](/document-ai/docs/processors-list)."]]
